Derek J. Lomas
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Derek J. Lomas has a total of 95 co-authors. Here are the top co-authors by number of shared publications:
- Lance A. Mynderse
Mayo Clinic
3 shared publications - Scott M. Thompson
Division of Vascular and Interventional Radiology, Department of Radiology, Mayo Clinic, Rochester, USA
3 shared publications - Akira Kawashima
Mayo Clinic Arizona
2 shared publications Christopher Favazza
Department of Medical Physics, Mayo Clinic, Rochester, Minnesota
2 shared publications- Daniel A. Adamo
Department of Radiology, Mayo Clinic, Rochester, USA
2 shared publications - Daniel S. Elliott
Department of Urology, Mayo Clinic, Rochester, Minnesota
2 shared publications
Latest Publications
- Open Access
- 17 October 2026
- Published by Springer Nature in Abdominal Radiology
- p. 1-8
Photon counting detector CTA for prostate arterial evaluation during the workup for prostate artery embolization
Purpose To evaluate small field of view (FOV) ultra-high spatial resolution pelvic photon-counting detector prostate CT angiograms (UHR PCD CTA) for identifying prostate arterial anatomy and extraprostatic anastomoses in patients undergoing workup for prostate artery embolization (PAE). Methods IRB-approved retrospective imaging and electronic medical record review of men who consecutively underwent UHR PCD CTA for PAE workup at our institution from 8/1/2022 to 12/1/2024. UHR PCD CTAs were evaluated by two vascular and interventional radiologists for main and accessory candidate prostatic arteries (PA/s) and anastomoses to penile, bladder, or rectal arteries. A main PA was defined as supplying the central gland and an accessory PA was defined as supplying the peripheral zone. PAs were categorized using the Carnevale classification. In patients who proceeded to PAE, digital subtraction angiograms (DSA) were retrospectively compared with UHR PCD CTA findings. Results UHR PCD CTA was obtained prior to PAE consultation for 50 men. We evaluated 100 pelvic sides, identifying 180 candidate PAs with a median of 3 candidate PAs per patient (mean, 3.6; range, 2â7). Overall, 56% (101/180) of candidate PAs were main and 44% (79/180) were accessory. Abridged Carnevale classifications: Type 5â21% (38/180). Eight accessory candidate PAs arose from the superior rectal artery. At least one candidate anastomosis arose from 40% (72/180) of candidate PAs. Of 95 total candidate anastomoses, 69 were extraprostatic (30 rectal, 20 bladder, 19 penile) and 26 were intraprostatic collaterals; 56% (53/95) arose from accessory and 44% (42/95) from main candidate PAs. Twenty-eight of 50 (56%) men underwent PAE, during which 72% (74/103) of available candidate PAs were selected, and prostatic supply was confirmed by DSA in 100% (74/74) of selected vessels. Conclusion Small FOV UHR PCD CTA is useful in identifying candidate PA supply, as well as potential extraprostatic anastomoses, providing an arterial roadmap for PAE planning. Candidate PA supply was confirmed by DSA in all selected vessels among patients who proceeded to PAE.
- 17 October 2026
- Published by Springer Nature in Abdominal Radiology
- p. 1-8
Open AccessPhoton counting detector CTA for prostate arterial evaluation during the workup for prostate artery embolization
Purpose To evaluate small field of view (FOV) ultra-high spatial resolution pelvic photon-counting detector prostate CT angiograms (UHR PCD CTA) for identifying prostate arterial anatomy and extraprostatic anastomoses in patients undergoing workup for prostate artery embolization (PAE). Methods IRB-approved retrospective imaging and electronic medical record review of men who consecutively underwent UHR PCD CTA for PAE workup at our institution from 8/1/2022 to 12/1/2024. UHR PCD CTAs were evaluated by two vascular and interventional radiologists for main and accessory candidate prostatic arteries (PA/s) and anastomoses to penile, bladder, or rectal arteries. A main PA was defined as supplying the central gland and an accessory PA was defined as supplying the peripheral zone. PAs were categorized using the Carnevale classification. In patients who proceeded to PAE, digital subtraction angiograms (DSA) were retrospectively compared with UHR PCD CTA findings. Results UHR PCD CTA was obtained prior to PAE consultation for 50 men. We evaluated 100 pelvic sides, identifying 180 candidate PAs with a median of 3 candidate PAs per patient (mean, 3.6; range, 2â7). Overall, 56% (101/180) of candidate PAs were main and 44% (79/180) were accessory. Abridged Carnevale classifications: Type 5â21% (38/180). Eight accessory candidate PAs arose from the superior rectal artery. At least one candidate anastomosis arose from 40% (72/180) of candidate PAs. Of 95 total candidate anastomoses, 69 were extraprostatic (30 rectal, 20 bladder, 19 penile) and 26 were intraprostatic collaterals; 56% (53/95) arose from accessory and 44% (42/95) from main candidate PAs. Twenty-eight of 50 (56%) men underwent PAE, during which 72% (74/103) of available candidate PAs were selected, and prostatic supply was confirmed by DSA in 100% (74/74) of selected vessels. Conclusion Small FOV UHR PCD CTA is useful in identifying candidate PA supply, as well as potential extraprostatic anastomoses, providing an arterial roadmap for PAE planning. Candidate PA supply was confirmed by DSA in all selected vessels among patients who proceeded to PAE.
- 9 March 2026
- Published by Springer Nature in Abdominal Radiology
- p. 1-10
Scrotal uptake on PSMA PET in prostate cancer: differentiating metastases from benign and physiologic findings using ultrasound correlation
Purpose To identify imaging and clinical features that aid in distinguishing metastatic lesions from benign or physiologic uptake on PSMA PET, using correlation with scrotal ultrasound. Methods This IRB-approved retrospective study included prostate cancer (PCa) patients who underwent PSMA PET-CT or PET-MRI between July 2021 and May 2025, with a scrotal ultrasound performed within 100 days of the scan. PSMA-avid lesions were assessed for uptake intensity (mild vs. intense), distribution (diffuse vs. focal), laterality (unilateral vs. bilateral), scan indication (initial staging vs. restaging), and SUVmax. Ultrasound findings were reviewed, and metastasis was confirmed by pathology or clinical consensus by three board-certified radiologists. Multivariable logistic regression using LASSO was performed to identify predictors of testicular or paratesticular metastasis. Results Of 114 patients (mean age 70), 24 (21%) demonstrated scrotal PSMA uptake. Focal intense uptake was seen in 6 patients (all unilateral), and mild focal uptake in 17 (13 bilateral, 4 unilateral). One patient showed mild diffuse bilateral uptake; no cases had intense diffuse uptake. Metastasis was confirmed in 5 patients, all of whom showed focal, unilateral, intense uptake. Among the 17 with mild focal uptake, 16 (94%) had no ultrasound abnormalities, consistent with physiologic epididymal uptake. Of 12 patients with ultrasound findings of epididymitis or orchitis, only 2 showed PSMA uptake (1 intense, 1 mild), while 10 had none. LASSO regression identified uptake intensity, SUVmax, PET indication, and laterality as significant predictors of metastasis. PET indication showed the strongest association with metastasis (coefficient = 4.60), followed by uptake intensity (3.97), SUVmax (1.37), and laterality (0.37), indicating a higher likelihood of metastasis in unilateral, intensely avid lesions, particularly in restaging settings. Conclusion Focal, unilateral, intense PSMA uptake, especially during restaging, was strongly associated with testicular or paratesticular metastasis. In contrast, mild or bilateral uptake without ultrasound abnormalities was typically physiologic. Combined assessment using PSMA PET and scrotal ultrasound improves diagnostic accuracy in evaluating scrotal findings in PCa patients.
- 9 March 2026
- Published by Springer Nature in Abdominal Radiology
- p. 1-10
Scrotal uptake on PSMA PET in prostate cancer: differentiating metastases from benign and physiologic findings using ultrasound correlation
Purpose To identify imaging and clinical features that aid in distinguishing metastatic lesions from benign or physiologic uptake on PSMA PET, using correlation with scrotal ultrasound. Methods This IRB-approved retrospective study included prostate cancer (PCa) patients who underwent PSMA PET-CT or PET-MRI between July 2021 and May 2025, with a scrotal ultrasound performed within 100 days of the scan. PSMA-avid lesions were assessed for uptake intensity (mild vs. intense), distribution (diffuse vs. focal), laterality (unilateral vs. bilateral), scan indication (initial staging vs. restaging), and SUVmax. Ultrasound findings were reviewed, and metastasis was confirmed by pathology or clinical consensus by three board-certified radiologists. Multivariable logistic regression using LASSO was performed to identify predictors of testicular or paratesticular metastasis. Results Of 114 patients (mean age 70), 24 (21%) demonstrated scrotal PSMA uptake. Focal intense uptake was seen in 6 patients (all unilateral), and mild focal uptake in 17 (13 bilateral, 4 unilateral). One patient showed mild diffuse bilateral uptake; no cases had intense diffuse uptake. Metastasis was confirmed in 5 patients, all of whom showed focal, unilateral, intense uptake. Among the 17 with mild focal uptake, 16 (94%) had no ultrasound abnormalities, consistent with physiologic epididymal uptake. Of 12 patients with ultrasound findings of epididymitis or orchitis, only 2 showed PSMA uptake (1 intense, 1 mild), while 10 had none. LASSO regression identified uptake intensity, SUVmax, PET indication, and laterality as significant predictors of metastasis. PET indication showed the strongest association with metastasis (coefficient = 4.60), followed by uptake intensity (3.97), SUVmax (1.37), and laterality (0.37), indicating a higher likelihood of metastasis in unilateral, intensely avid lesions, particularly in restaging settings. Conclusion Focal, unilateral, intense PSMA uptake, especially during restaging, was strongly associated with testicular or paratesticular metastasis. In contrast, mild or bilateral uptake without ultrasound abnormalities was typically physiologic. Combined assessment using PSMA PET and scrotal ultrasound improves diagnostic accuracy in evaluating scrotal findings in PCa patients.
- 1 November 2025
- Published by Wolters Kluwer Health in Urology Practice
- Vol. 12  (6) , 651
Editorial Commentary
- 1 November 2025
- Published by Wolters Kluwer Health in Urology Practice
- Vol. 12  (6) , 651
Editorial Commentary
- Open Access
- 15 July 2025
- Published by MDPI AG in Diagnostics
- Vol. 15  (14) , 1781
In-Bore MRI-Guided Ureteral Stent Placement During Prostate Cancer CryoablationâA Case Series
Introduction: Ureteral stents are widely used in the specialty of urology to preserve renal function and provide ureteral patency in cases of urolithiasis, strictures, malignancy, and trauma. This paper presents a novel application of prophylactic ureteral stents deployed under MRI-guidance for ureteral protection in the setting of in-bore salvage cryoablation therapy for recurrent and metastatic prostate cancer. This is the first known case series of ureteral stent placement using near real-time MRI. Materials and Methods: A retrospective chart review was performed for all patients who underwent MRI-guided ureteral stent placement prior to in-bore cryoablation therapy from 2021 to 2022. Each case was managed by an interdisciplinary team of urologists and interventional radiologists. Preoperative and postoperative data were collected for descriptive analysis. Physics safety testing was conducted on the cystoscope and viewing apparatus prior to its implementation for stent deployment. Results: A total of seven males, mean age 73.4 years (range 65â81), underwent successful prophylactic, cystoscopic MRI-guided ureteral stent placement prior to cryoablation therapy of their prostate cancer. No intraoperative complications occurred. A Grade 2 postoperative complication of pyelonephritis and gross hematuria following stent removal occurred in one case. The majority of patients were discharged the same day as their procedure. Conclusions: This case series demonstrates the feasibility of in-bore cystoscopic aided MRI guidance for ureteral stent placement. Ureteral stents can be used to increase the safety margin of complex cryoablation treatments close to the ureter. Furthermore, by following the meticulous MRI safety protocols established by MRI facility safety design guidelines, MRI conditional tools can aid therapy in the burgeoning interventional MRI space.
- 15 July 2025
- Published by MDPI AG in Diagnostics
- Vol. 15  (14) , 1781
Open AccessIn-Bore MRI-Guided Ureteral Stent Placement During Prostate Cancer CryoablationâA Case Series
Introduction: Ureteral stents are widely used in the specialty of urology to preserve renal function and provide ureteral patency in cases of urolithiasis, strictures, malignancy, and trauma. This paper presents a novel application of prophylactic ureteral stents deployed under MRI-guidance for ureteral protection in the setting of in-bore salvage cryoablation therapy for recurrent and metastatic prostate cancer. This is the first known case series of ureteral stent placement using near real-time MRI. Materials and Methods: A retrospective chart review was performed for all patients who underwent MRI-guided ureteral stent placement prior to in-bore cryoablation therapy from 2021 to 2022. Each case was managed by an interdisciplinary team of urologists and interventional radiologists. Preoperative and postoperative data were collected for descriptive analysis. Physics safety testing was conducted on the cystoscope and viewing apparatus prior to its implementation for stent deployment. Results: A total of seven males, mean age 73.4 years (range 65â81), underwent successful prophylactic, cystoscopic MRI-guided ureteral stent placement prior to cryoablation therapy of their prostate cancer. No intraoperative complications occurred. A Grade 2 postoperative complication of pyelonephritis and gross hematuria following stent removal occurred in one case. The majority of patients were discharged the same day as their procedure. Conclusions: This case series demonstrates the feasibility of in-bore cystoscopic aided MRI guidance for ureteral stent placement. Ureteral stents can be used to increase the safety margin of complex cryoablation treatments close to the ureter. Furthermore, by following the meticulous MRI safety protocols established by MRI facility safety design guidelines, MRI conditional tools can aid therapy in the burgeoning interventional MRI space.
- 3 June 2025
- Published by Springer Nature in Abdominal Radiology
- Vol. 50  (12) , 5948-5962
Localization of PSMA-avid lesions on PSMA PET-CT on prostate MRI in patients with PI-RADS 3
Purpose To localize PSMA-avid lesions identified by PSMA PET-CT on separately performed prostate MRI and evaluate imaging findings on fused PET-CT/MRI. Methods Patients without prior history of clinically significant prostate cancer (csPCa: Gleason score [GS] 3 + 4 or higher) who had (1) PI-RADS 3 on prostate MRI between 2021 and 2023, (2) MRI/US fusion targeted biopsy for PI-RADS 3 lesion(s) and systemic biopsy, and (3) subsequent PSMA PET-CT were identified. PSMA PET-CT images were fused onto prostate MRI. PI-RADS 3 lesions were categorized by PRIMARY score. Discordant PSMA-avid lesions with PRIMARY scores 2â5 outside of PI-RADS 3 lesions were identified. Fisherâs exact test was used to compare the proportion of csPCa on targeted biopsy between PSMA-positive (PRIMARY score 3 or more) and PSMA-negative (PRIMARY score 1 or 2) PI-RADS 3 lesions. P < 0.05 was considered statistically significant. Results 30 patients (mean age 67 years) with 38 PI-RADS 3 lesions were identified. 29 patients had csPCa, and one patient had GS 6. 22 PI-RADS 3 lesions were PSMA-positive (PRIMARY score 3 or more), of which 18 (81.8%) were csPCa on targeted biopsy; 16 PI-RADS 3 lesions were PSMA-negative, of which 4 (25.0%) were csPCa on targeted biopsy (p < 0.001). Sensitivity and specificity for the presence of csPCa on targeted biopsy was 81.8% and 75.0%. Out of 30 patients, 12 (40.0%) had 16 discordant PSMA-avid (PRIMARY score 2 or more) lesions and 10 patients had 10 discordant PSMA-positive lesions outside PI-RADS 3 lesions In 5 of those 12 (41.7%), discordant PSMA-avid lesions had higher PRIMARY score than PI-RADS 3 lesions. In 4 of those 5 (80%), systemic biopsy showed higher GS than targeted biopsy. Conclusion Lesion-level analysis showed PSMA-positive PI-RADS 3 lesions had higher probability of csPCa than PSMA-negative PI-RADS 3 lesions. Discordant PSMA-avid lesions with higher PRIMARY score than that of PI-RADS 3 lesions often represented another more aggressive focus not initially identified on MRI.
- 3 June 2025
- Published by Springer Nature in Abdominal Radiology
- Vol. 50  (12) , 5948-5962
Localization of PSMA-avid lesions on PSMA PET-CT on prostate MRI in patients with PI-RADS 3
Purpose To localize PSMA-avid lesions identified by PSMA PET-CT on separately performed prostate MRI and evaluate imaging findings on fused PET-CT/MRI. Methods Patients without prior history of clinically significant prostate cancer (csPCa: Gleason score [GS] 3 + 4 or higher) who had (1) PI-RADS 3 on prostate MRI between 2021 and 2023, (2) MRI/US fusion targeted biopsy for PI-RADS 3 lesion(s) and systemic biopsy, and (3) subsequent PSMA PET-CT were identified. PSMA PET-CT images were fused onto prostate MRI. PI-RADS 3 lesions were categorized by PRIMARY score. Discordant PSMA-avid lesions with PRIMARY scores 2â5 outside of PI-RADS 3 lesions were identified. Fisherâs exact test was used to compare the proportion of csPCa on targeted biopsy between PSMA-positive (PRIMARY score 3 or more) and PSMA-negative (PRIMARY score 1 or 2) PI-RADS 3 lesions. P < 0.05 was considered statistically significant. Results 30 patients (mean age 67 years) with 38 PI-RADS 3 lesions were identified. 29 patients had csPCa, and one patient had GS 6. 22 PI-RADS 3 lesions were PSMA-positive (PRIMARY score 3 or more), of which 18 (81.8%) were csPCa on targeted biopsy; 16 PI-RADS 3 lesions were PSMA-negative, of which 4 (25.0%) were csPCa on targeted biopsy (p < 0.001). Sensitivity and specificity for the presence of csPCa on targeted biopsy was 81.8% and 75.0%. Out of 30 patients, 12 (40.0%) had 16 discordant PSMA-avid (PRIMARY score 2 or more) lesions and 10 patients had 10 discordant PSMA-positive lesions outside PI-RADS 3 lesions In 5 of those 12 (41.7%), discordant PSMA-avid lesions had higher PRIMARY score than PI-RADS 3 lesions. In 4 of those 5 (80%), systemic biopsy showed higher GS than targeted biopsy. Conclusion Lesion-level analysis showed PSMA-positive PI-RADS 3 lesions had higher probability of csPCa than PSMA-negative PI-RADS 3 lesions. Discordant PSMA-avid lesions with higher PRIMARY score than that of PI-RADS 3 lesions often represented another more aggressive focus not initially identified on MRI.
- 1 December 2024
- Published by Springer Nature
- p. 469-474
Salvage Treatment Following Focal Therapy
- 30 November 2024
- Published by Wiley in BJU International
- Vol. 135  (4) , 668-674
Inâfield prostate cancer recurrence following radical prostatectomy and salvage radiation
Objective: To define the natural history, patterns of recurrence and treatment modalities for local prostate cancer (PCa) recurrence following radical prostatectomy (RP) and radiation therapy (RT), and to investigate factors that could predict metastasisâfree survival (MFS) in this unique patient population. Methods: We queried a prospectively maintained PCa registry to identify men developing inâfield recurrence (IFR) following RP and RT from 2008 to 2021 at a single institution. IFR was defined as biopsyâproven recurrent PCa or the presence of persistent positron emission tomographyâavid lesions in the prior radiation field without evidence of metastasis. Cox regression was conducted to determine predictors of MFS. KaplanâMeier methods were used to calculate MFS, cancerâspecific survival (CSS) and overall survival (OS) for patients in three primary therapy categories: cryoablation, androgen deprivation therapy (ADT) alone, and surveillance. Results: Of 4575 patients from our registry, 108 (2.3%) with IFR were identified. The median (interquartile range [IQR]) time to IFR from salvage treatment was 78 (50â126) months. A total of 29 patients (26%) were managed with cryoablation, 23 (21%) with ADT, and 28 (25%) with surveillance. The median (IQR) followâup was 76 (48â100) months. There were no statistically significant differences in MFS ( P = 0.67) or OS ( P = 0.07) among the three primary treatment cohorts. Patients treated with ADT or cryoablation had longer CSS compared to patients managed with surveillance ( P = 0.047). Conclusions: We found that IFR may present years after completion of primary treatment for PCa. While curative management strategies may be attempted, local and distant metastatic recurrence is common and often requires systemic therapy.
- 30 November 2024
- Published by Wiley in BJU International
- Vol. 135  (4) , 668-674
Inâfield prostate cancer recurrence following radical prostatectomy and salvage radiation
Objective: To define the natural history, patterns of recurrence and treatment modalities for local prostate cancer (PCa) recurrence following radical prostatectomy (RP) and radiation therapy (RT), and to investigate factors that could predict metastasisâfree survival (MFS) in this unique patient population. Methods: We queried a prospectively maintained PCa registry to identify men developing inâfield recurrence (IFR) following RP and RT from 2008 to 2021 at a single institution. IFR was defined as biopsyâproven recurrent PCa or the presence of persistent positron emission tomographyâavid lesions in the prior radiation field without evidence of metastasis. Cox regression was conducted to determine predictors of MFS. KaplanâMeier methods were used to calculate MFS, cancerâspecific survival (CSS) and overall survival (OS) for patients in three primary therapy categories: cryoablation, androgen deprivation therapy (ADT) alone, and surveillance. Results: Of 4575 patients from our registry, 108 (2.3%) with IFR were identified. The median (interquartile range [IQR]) time to IFR from salvage treatment was 78 (50â126) months. A total of 29 patients (26%) were managed with cryoablation, 23 (21%) with ADT, and 28 (25%) with surveillance. The median (IQR) followâup was 76 (48â100) months. There were no statistically significant differences in MFS ( P = 0.67) or OS ( P = 0.07) among the three primary treatment cohorts. Patients treated with ADT or cryoablation had longer CSS compared to patients managed with surveillance ( P = 0.047). Conclusions: We found that IFR may present years after completion of primary treatment for PCa. While curative management strategies may be attempted, local and distant metastatic recurrence is common and often requires systemic therapy.
- 22 June 2024
- Published by Elsevier in Journal of Vascular and Interventional Radiology
- Vol. 35  (10) , 1474-1480
Magnetic Resonance ImagingâGuided Cryoablation of Prostate Cancer Lymph Node Metastasis
- 22 June 2024
- Published by Elsevier in Journal of Vascular and Interventional Radiology
- Vol. 35  (10) , 1474-1480
Magnetic Resonance ImagingâGuided Cryoablation of Prostate Cancer Lymph Node Metastasis
- Open Access
A Description and Safety Overview of Irreversible Electroporation for Prostate Tissue Ablation in Intermediate-Risk Prostate Cancer Patients: Preliminary Results from the PRESERVE Trial
The PRESERVE study (NCT04972097) aims to evaluate the safety and effectiveness of the NanoKnife System to ablate prostate tissue in patients with intermediate-risk prostate cancer (PCa). The NanoKnife uses irreversible electroporation (IRE) to deliver high-voltage electrical pulses to change the permeability of cell membranes, leading to cell death. A total of 121 subjects with organ-confined PCa ⤠T2c, prostate-specific antigens (PSAs) ⤠15 ng/mL, and a Gleason score of 3 + 4 or 4 + 3 underwent focal ablation of the index lesion. The primary endpoints included negative in-field biopsy and adverse event incidence, type, and severity through 12 months. At the time of analysis, the trial had completed accrual with preliminary follow-up available. Demographics, disease characteristics, procedural details, PSA responses, and adverse events (AEs) are presented. The median (IQR) age at screening was 67.0 (61.0â72.0) years and Gleason distribution 3 + 4 (80.2%) and 4 + 3 (19.8%). At 6 months, all patients with available data (n = 74) experienced a median (IQR) percent reduction in PSA of 67.6% (52.3â82.2%). Only ten subjects (8.3%) experienced a Grade 3 adverse event; five were procedure-related. No Grade ⼠4 AEs were reported. This study supports prior findings that IRE prostate ablation with the NanoKnife System can be performed safely. Final results are required to fully assess oncological, functional, and safety outcomes.
Open AccessA Description and Safety Overview of Irreversible Electroporation for Prostate Tissue Ablation in Intermediate-Risk Prostate Cancer Patients: Preliminary Results from the PRESERVE Trial
The PRESERVE study (NCT04972097) aims to evaluate the safety and effectiveness of the NanoKnife System to ablate prostate tissue in patients with intermediate-risk prostate cancer (PCa). The NanoKnife uses irreversible electroporation (IRE) to deliver high-voltage electrical pulses to change the permeability of cell membranes, leading to cell death. A total of 121 subjects with organ-confined PCa ⤠T2c, prostate-specific antigens (PSAs) ⤠15 ng/mL, and a Gleason score of 3 + 4 or 4 + 3 underwent focal ablation of the index lesion. The primary endpoints included negative in-field biopsy and adverse event incidence, type, and severity through 12 months. At the time of analysis, the trial had completed accrual with preliminary follow-up available. Demographics, disease characteristics, procedural details, PSA responses, and adverse events (AEs) are presented. The median (IQR) age at screening was 67.0 (61.0â72.0) years and Gleason distribution 3 + 4 (80.2%) and 4 + 3 (19.8%). At 6 months, all patients with available data (n = 74) experienced a median (IQR) percent reduction in PSA of 67.6% (52.3â82.2%). Only ten subjects (8.3%) experienced a Grade 3 adverse event; five were procedure-related. No Grade ⼠4 AEs were reported. This study supports prior findings that IRE prostate ablation with the NanoKnife System can be performed safely. Final results are required to fully assess oncological, functional, and safety outcomes.
- Open Access
- 7 March 2024
- Published by Frontiers Media SA in Frontiers in Artificial Intelligence
- Vol. 7, 1375482
Automated surgical step recognition in transurethral bladder tumor resection using artificial intelligence: transfer learning across surgical modalities
Objective: Automated surgical step recognition (SSR) using AI has been a catalyst in the âdigitizationâ of surgery. However, progress has been limited to laparoscopy, with relatively few SSR tools in endoscopic surgery. This study aimed to create a SSR model for transurethral resection of bladder tumors (TURBT), leveraging a novel application of transfer learning to reduce video dataset requirements.Materials and methods: Retrospective surgical videos of TURBT were manually annotated with the following steps of surgery: primary endoscopic evaluation, resection of bladder tumor, and surface coagulation. Manually annotated videos were then utilized to train a novel AI computer vision algorithm to perform automated video annotation of TURBT surgical video, utilizing a transfer-learning technique to pre-train on laparoscopic procedures. Accuracy of AI SSR was determined by comparison to human annotations as the reference standard.Results: A total of 300 full-length TURBT videos (median 23.96 min; IQR 14.13â41.31 min) were manually annotated with sequential steps of surgery. One hundred and seventy-nine videos served as a training dataset for algorithm development, 44 for internal validation, and 77 as a separate test cohort for evaluating algorithm accuracy. Overall accuracy of AI video analysis was 89.6%. Model accuracy was highest for the primary endoscopic evaluation step (98.2%) and lowest for the surface coagulation step (82.7%).Conclusion: We developed a fully automated computer vision algorithm for high-accuracy annotation of TURBT surgical videos. This represents the first application of transfer-learning from laparoscopy-based computer vision models into surgical endoscopy, demonstrating the promise of this approach in adapting to new procedure types.
- 7 March 2024
- Published by Frontiers Media SA in Frontiers in Artificial Intelligence
- Vol. 7, 1375482
Open AccessAutomated surgical step recognition in transurethral bladder tumor resection using artificial intelligence: transfer learning across surgical modalities
Objective: Automated surgical step recognition (SSR) using AI has been a catalyst in the âdigitizationâ of surgery. However, progress has been limited to laparoscopy, with relatively few SSR tools in endoscopic surgery. This study aimed to create a SSR model for transurethral resection of bladder tumors (TURBT), leveraging a novel application of transfer learning to reduce video dataset requirements.Materials and methods: Retrospective surgical videos of TURBT were manually annotated with the following steps of surgery: primary endoscopic evaluation, resection of bladder tumor, and surface coagulation. Manually annotated videos were then utilized to train a novel AI computer vision algorithm to perform automated video annotation of TURBT surgical video, utilizing a transfer-learning technique to pre-train on laparoscopic procedures. Accuracy of AI SSR was determined by comparison to human annotations as the reference standard.Results: A total of 300 full-length TURBT videos (median 23.96 min; IQR 14.13â41.31 min) were manually annotated with sequential steps of surgery. One hundred and seventy-nine videos served as a training dataset for algorithm development, 44 for internal validation, and 77 as a separate test cohort for evaluating algorithm accuracy. Overall accuracy of AI video analysis was 89.6%. Model accuracy was highest for the primary endoscopic evaluation step (98.2%) and lowest for the surface coagulation step (82.7%).Conclusion: We developed a fully automated computer vision algorithm for high-accuracy annotation of TURBT surgical videos. This represents the first application of transfer-learning from laparoscopy-based computer vision models into surgical endoscopy, demonstrating the promise of this approach in adapting to new procedure types.